Package: lirc Version: 0.9.0~pre1-1 Severity: normal Hi,
I was trying to get lirc running, but for some reason the /var/run/lirc directory showed up empty after boot. Lircd was running, but the socket and pidfile were not there. After a morning of debugging I found out what was happening, but I'm not sure whose "fault" this is or what is the best way to fix it, nor why this does not affect all users. The system I'm trying this on is a highly customized Debian-live setup, but I don't think that the behaviour I'm about to describe is limited to this system. This system uses an on-board CIR controller (connected internally through PCI I think), so perhaps this bug doesn't trigger for USB devices, only for PCI devices and that's why nobody noticed it until now? Anyway, here's what happens for me: - During system boot, /etc/rcS.d/S03udev makes udev synthesize initial hotplug add events. - Udev responds to the add event for the CIR controller by calling /etc/init.d/lirc restart, which starts lircd. - /etc/rcS.d/S13mountall-bootclean.sh then runs, and cleans out all files in /var/run/lirc (leaving the directory to exist, I think). - /etc/rc2.d/S19lirc runs, but lircd doesn't start because it's already running. The end result is a system with a running lircd, but no way to actually reach it. I'll be working around this problem by removing the udev rules file, but I'm happy to help out with further details or by testing proposed fixes. Gr. Matthijs -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
